Hand Written Signature Image Analysis and Recognition

Tai-Hoon Kim

Department of Multimedia, Hannam University, Daejeon, Korea.


Handwritten Signature Identification is a classical work area in the line of Computer Science and Technology since last few years. Various new techniques of Image Analysis also attracting the Computer Scientists as well. Firstly, Pixel clustering is used to transform the signature image into bi-color image. Then secondly, instead of considering the whole image, only signature area is extracted. Thirdly, by using Image scaling technique the signature image resized along the coordinate directions. As different techniques are used to subsample (image after transformation) which will be discussed in turn. Fourthly, a different technique is used for thinning to reduce the threshold output of an edge detector algorithm is used to lines of a single pixel thickness. In this paper I propose the above mentioned series of techniques as the preprocessing analysis part of Handwritten Signature Recognition.

Keywords: Skeletonization, Scaling, ITA (Image Thinning Algorithm), ROI (Region of Interest).

     Back to TOC